1Liver Unit, Department of Medical Sciences, University of Torino, 10126 Turin, Italy.
2Gastroenterology Unit, Department of General and Specialistic Medicine, A.O.U. Città della Salute e della Scienza-Molinette Hospital, 10126 Turin, Italy.
Patients with hepatitis C virus (HCV)-related cirrhosis treated with direct-acting antivirals (DAA) are still at risk of developing hepatocellular carcinoma (HCC). We investigated the accuracy of non-invasive scoring systems (NSS) for the prediction of de novo HCC development in patients treated with DAA on long-term follow-up (FU). We analyzed data from 575 consecutive patients with cirrhosis and no history of HCC who achieved a sustained virologic response (SVR) to DAA therapy. NSS (i.e., Forns index, APRI, FIB-4, ALBI, and aMAP) were calculated at 3 months after the end of therapy. Performance for de novo HCC prediction was evaluated in terms of area under the curve (AUC) and Harrell's C-index. During a median FU of 44.9 (27.8-58.6) months, 57 (9.9%) patients developed de novo HCC. All five NSS were associated with the risk of de novo HCC. At multivariate analysis, only the ALBI score resulted in being significantly and independently associated with de novo HCC development (adjusted hazard ratio = 4.91, 95% CI 2.91-8.28, p < 0.001). ALBI showed the highest diagnostic accuracy for the detection of de novo HCC at 1-, 3-, and 5-years of FU, with AUC values of 0.81 (95% CI 0.78-0.85), 0.71 (95% CI 0.66-0.75), and 0.68 (95% CI 0.59-0.76), respectively. Consistently, the best predictive performance assessed by Harrell's C-statistic was observed for ALBI (C-index = 0.70, 95% CI 0.62-0.77). ALBI score may represent a valuable and inexpensive tool for risk stratification and the personalization of an HCC surveillance strategy for patients with cirrhosis and previous history of HCV infection treated with DAA.